Waste Management 22 (2002) 945–949 www.elsevier.com/locate/wasman
Vitrification of electric arc furnace dusts M. Pelino*, A. Karamanov, P. Pisciella, S. Crisucci, D. Zonetti Department of Chemistry, Chemical Engineering and Materials, University of L’Aquila, Monteluco di Roio, 67040 L’Aquila, Italy Accepted 13 June 2002
Abstract Electric arc furnace baghouse dust (EAFD), a waste by-product of the steelmaking process, contains the elements that are volatilized from the charge during the melting (Cr, Pb, Zn, Cu and Cd). The results of leaching tests show that the concentration of these elements exceeds the regulatory limits. Consequently, EAFD cannot be disposed of in ordinary landfill sites without stabilization of the heavy metals. In this work, the vitrification of EAFD, from both carbon and stainless steel productions, were studied. The vitrification process was selected as the inertizing process because it permits the immobilization of the hazardous elements in the glass network and represents an environmentally acceptable method for the stabilization of this waste. Glasses of various compositions were obtained by mixing EAFD with glass cullet and sand. The EAFD and the glass products were characterized by DTA, TG, X-ray analysis and by the TCLP test. The results show that the stability of the product is influenced by the glass structure, which mainly depends on the Si/O ratio. Secondary crystallization heat-treatment were carried out on some samples. The results highlighted the formation of spinel phases, which reduced the chemical durability in acid media. The possibility to recover Zn from carbon steel production EAFD was investigated and about 60–70% of metal recovery was obtained. The resulting glass show higher chemical stability than glasses obtained without metal recovery. # 2002 Elsevier Science Ltd. All rights reserved.
1. Introduction EAFD is generated by the electric arc steelmaking process. During melting certain elements volatilize and, after oxidation and cooling, are collected in filters. Approximately 1–2% of the charge is converted into EAFD (Zunkel, 1997). About 650–700 kt/year of EAFD are produced in USA and 1 million t/year in EU. About 200 kt/year of EAFD are generated in Italy from carbon and stainless steel production. The disposal of this waste has become a serious problem in recent years due to the fact that the greater availability of steel scrap, which represents a big part of the charge, has increased the production of steel from electric arc furnaces. Considering the current rate of growth it is expected that the world production of EAFD will reach about 5 millions t/year in 2005 (Goodwill and Schmitt, 1994). The dust contains elements such us Pb, Cr, Zn, Cd, Cu, whose
* Corresponding author. Tel.: +39-0862-434224; fax: +39-0862434233. E-mail address:
[email protected] (M. Pelino).
solubility in leaching media exceeds the environmental regulatory limits in EU and USA. Several treatment technologies have been developed and can be divided into two main groups: processes which recover the metals and stabilization processes. The greater part of EAFD presently is treated by a high temperature metal recovery (HTMR) process and successive condensation. This technology allows the recovery of elements with high vapor pressure, such as Zn and Pb, and produces an iron rich residue which can be recycled as part of the furnace charge. This process is efficient for the treatment of dust generated in carbon steel production when the percentage of Zn and Pb is greater than 15–20% (Zunkel, 1997; Goodwill and Schmitt, 1994). The amount of Zn and Pb contained in the dust can be increased by a direct recycling of EAFD in the furnace, if the steel quality is not influenced (Evans and Hogan, 1987; Mankins and Shields, 1996; McCrea and Pikles, 1995). Metal recovery is also accomplished by several hydrometallurgical processes. When metal recovery is not profitable a stabilization process in a cement matrix can be used (Lynn, 1988) as well as vitrification, which is acquiring importance as
0956-053X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved. PII: S0956-053X(02)00080-6
946
M. Pelino et al. / Waste Management 22 (2002) 945–949
the ultimate mean for the immobilization of the heavy metals contained in the EAFD (Ionescu et al., 1995; Gress and Sarko, 1993; Ek and Schlobohm, 1993; Aota et al., 1995). In this work results of vitrification of two different EAFD, from stainless and carbon steel respectively, are presented. By mixing glass cullet, sand and EAFD in different ratios several types of glasses were obtained and characterized. The conditions to obtain a partial recovery of Zn from carbon steel EAFD were also investigated.
2. Experimental The EAFD studied in this work are generated in two different plants: for stainless steel (labeled EAFD 1) and carbon steel production (labeled EAFD 2) respectively. Leaching tests in 0.5 M acetic acid showed that both samples exceed the environmental regulatory limits; particularly for Cr in EAFD 1 and for Zn in EAFD 2. Dust chemical analysis was carried out by a EDXRF equipped with a 50 W Rh anode X-Ray tube. The densities was measured by a helium-displacement AccuPyc 1330 pycnometer, the particle size distribution was determined by MALVERN series 2600 laser diffraction technique and the specific surface by ASAP 2000 nitrogen porosimeter. The crystalline phases were analyzed by X-ray diffractometer (XRD), Philips 1830 utilizing a CuKa radiation. The thermal characterization of EAFD up to 1500 C was investigated by simultaneous thermal analysis (STA), by a NETSCH STA 409 analyzer. Different batches were prepared by mixing EAFD 1 or EAFD 2 with glass cullet and sand. Melting was carried out by a super Kantal-kiln at temperatures ranging from 1450 to 1500 C for 1–2 h using alumina crucibles, followed by quenching in water and steel mould. In EAFD 2 batches, different amount of carbon (from 0 to 10%) were added in order to enhance the zinc vaporization during a preliminary heat treatment at 1200 C for 5 h in air or nitrogen atmospheres. Glasses were characterized by EDXRF, XRD and STA and the chemical durability by leaching in acid acetic 0.5 M (TCLP).
3. Results and discussion 3.1. Dust characterization The chemical compositions of the EAFD are shown in Table 1 together with the composition of the glass cullet. Besides Fe oxides, EAFD 1 is characterized by the presence of Cr and Ca oxides while EAFD 2 by
Table 1 Chemical composition (wt.%) of EAFD 1, EAFD 2 and glass cullet
SiO2 Al2O3 Cr2O3 Fe2O3 CaO MgO MnO ZnO CuO Na2 O K2 O NiO PbO
EAFD 1
EAFD 2
Glass-cullet
4.4 1.48 15.85 24.28 20.69 9.6 5.05 7.57 0.41 6.62 1.76 1.8 0.49
5.94 0.65 1.12 52.82 7.50 5.21 5.33 13.80 0.18 0.91 1.01 0.08 0.48
72.53 0.51 – 0.11 8.89 3.95 – – – 13.81 0.20 – –
ZnO. XRD analysis showed that in EAFD 1 the major crystalline phases are the magnetite–chromite spinel, the calcium carbonate and ZnO. In EAFD 2 sample magnetite and ZnO are the major phases with traces of FeS. The particle size is very fine. Laser diffraction analyses show that EAFD 1 and EAFD 2 have 34 and 65% of particle size less than 10 mm, respectively; the specific surface resulted as 4.36 0.10 m2/g for EAFD 1 and 4.82 0.10 m2/g for EAFD 2. Densities are 3.740 0.002 and 4.257 0.002 g/cm3, respectively. TG-DTA up to 1000 C highlighted that weight loss for EAFD 1 is about 6%, due to CaCO3 decomposition (600–800 C), while is about 2% for EAFD 2. Above this temperature, weight losses are also observed due to the volatilization of Pb and Zn, accompanied by an endothermic effect in the DTA trace. This effect is more evident in EAFD 2 trace because of the higher zinc content. Similar results are reported by other authors (Aota et al., 1995). 3.2. Vitrification and products characterization Using EAFD 1 and glass cullet in the ratios 25/75, 40/ 60 and 50/50 three batches were prepared. The compositions, labeled as G1–25, G1–40, G1–50 are reported in Table 2. After melting, the chemical stability of the as quenched frit was tested by the TCLP. The results are presented in Table 3 while in Fig. 1, the concentrations of Cu, Al, Pb and Zn are plotted as a function of the percentage of EAFD 1 in the batch. In Fig. 2, the Si concentration in the leaching solution is plotted as function of the SiO2 percentage in the glass. It is evident that G1–25, G1–40 are stable glasses while G1–50 shows solubility in 0.5 M acid acetic. The chemical stability is influenced by the glass structure which mainly depends on the Si/O ratio. When this ratio is less than 0.33 there are no conditions for the formation of a continuos Si–O–Si network and the
947
M. Pelino et al. / Waste Management 22 (2002) 945–949
Fig. 1. Cu, Al, Pb and Zn concentration (mg/l) in the leaching test solution as a function of wt.% EAFD 1 in the batch.
resulting glass structure is not chemically stable (Appen, 1974). The Si/O ratio for the G1–25, G1–40, G1–50 glasses are reported in Table 2 and in Fig. 2. With the aim of maximizing the EAFD 1 content in the glass, part of the glass cullet was replaced by sand. Using 45% of EAFD 1, 27.5% of glass cullet and 27.5% of sand, the glass labeled G1–45 was obtained. The melting temperature was increased up to 1500 C for 2 h due to the higher silica percentage. The results of G1–45 leaching test are also reported in Table 3. The chemical stability of this glass is very high and comparable with the low EAFD content composition G1–25. Another aspect influencing the chemical durability of an amorphous material is crystallization. In commercial glass-ceramics, crystallization usually improves the chemical durability (Strnad, 1986) due to the formation of stable crystal phases and an increasing of SiO2 content in the residual glass. However, in particular systems, Table 2 Chemical composition (wt.%) of the investigated glasses
SiO2 Al2O3 Cr2O3 Fe2O3 CaO MgO MnO ZnO CuO Na2 O K2 O NiO PbO Si/O
G1–25
G1–40
G1–50
G1–45
G2–45
G2–50
56.15 0.73 3.79 5.89 11.71 5.3 1.32 1.81 0.1 12.07 0.57 0.43 0.12
46.12 0.88 6.15 9.48 13.47 6.14 1.96 2.94 0.16 11.01 0.81 0.7 0.19
39.4 0.99 7.74 11.87 14.64 6.7 2.46 3.69 0.2 10.31 0.96 0.88 0.24
50.57 0.84 6.9 10.62 11.48 5.3 2.23 3.32 0.18 6.76 0.84 0.79 0.22
46.9 0.44 0.48 25.65 6.48 3.9 2.52 6.52 0.58 5.73 0.74 0.03 0.22
45.82 0.37 0.58 29.52 6.44 2.89 2.85 2.81 0.07 4.50 0.30 0.03 0.02
0.35
0.30
0.27
0.33
0.32
0.31
Fig. 2. Si concentration (mg/l) in the leaching test solution and Si/O ratio as a function of SiO2 mol% in batch and wt.% EAFD 1 in the batch.
such as Li2O.SiO2, the crystal phase has lower stability than the parent glass. In our previous work (Pisciella et al., 2001), the chemical durability of Jarosite iron-rich glasses and glass-ceramics was investigated. In that case the chemical durability of the crystallized product depended on the ratio between the magnetite and pyroxene crystal phases. When pyroxene was the major phase, crystallization increased the chemical durability in both acid and basic conditions, while when the magnetite/pyroxene ratio was near one no improvements was noticed. In EAFD-1 glasses the influence of the crystallization on the chemical durability was investigated using the G1–50 composition, which showed the highest crystallization tendency. The crystallization heat-treatment was carried out for 2 h at 650 C. The XRD spectra reports 7–10% of cromite–spinel phase in parent glass. It must be pointed out that the term glass is improperly used, because, due to the high Cr, Zn and Fe content, EAFD compositions have a high tendency to crystallize even in drastic quenching conditions. The glass-cera-
Table 3 TCLP leaching test results for the EAFD 1 glasses (mg/l) Elements
G1–25
G1–40
G1–45
G1–50
GC1–50
Italian limits
Si Cr Al Ni Pb Cu Zn
0.23
